Evaluation of Corrosion Resistant Alloys as Construction Material for Acidic Geothermal Wells James Nogara1* and Sadiq J. Zarrouk2 1 2
Energy Development Corporation (EDC), One Corporate Center, Ortigas Center, Pasig City, Philippines
Department of Engineering Science, the University of Auckland, Private Bag 92019, Auckland, New Zealand *
[email protected]
Keywords: geothermal corrosion, acid fluid, corrosion resistant alloys, acid-sulfate chloride, casing, wellbore. ABSTRACT A number of geothermal fields around the world have reported excessive corrosion in wellbore and wellhead appurtenances constructed from standard carbon steel material due to acidic fluids (Keserovic and Bäßler, 2013; K. Lichti et al., 2010; Mundhenk, et al., 2013; Sanada et al., 2000, Villa, et al., 2000; Abe, 1993; Amend and Yee, 2013). This work focuses on two types of acidic geothermal fluid: (1) acid chlorine + (SO4=) type and (2) acid sulfate + (chloride) type. This report shows the damage to carbon steel piping material exposed to acid sulfate + (chloride) fluid with pH 2. The guideline also includes corrosion inhibition and pH modification aspects of corrosion control. No alloy material is recommended for use below pH 2. Finding suitable materials for handling acid geothermal fluids offers substantial economic incentive. Some very productive geothermal wells drilled lately in Sibayak and Lahendong in Indonesia (Keserovic and Bäßler, 2013) and recently in well IDDP1 in Iceland (Ármannsson et al., 2014) has been producing highly corrosive acid fluids. 1.2 Acidic Geothermal Fluids Sanada et al (1997) summarized the published chemistry data from acid wells from Japan, Philippines and Salton Sea, USA. Other published acid fluid chemistries are available in Einarsson, et al, 2010; Marini et al., 2003, Gherardi, et al., 2005; Abe, 1993; Mundhenk, et al., 2012, Koestono, 2010; Armansson, et al., 2014; and Truesdell, 1991). Three types of acid fluid were identified from geothermal well 2 New Zealand Geothermal Workshop 2014 Proceedings 24 – 26 November 2014 Auckland, New Zealand
discharges that have caused damage to geothermal wells and pipelines: (1) acid CI-SO4 waters with excess chloride, (2) HCl in superheated steam, and (3) acid SO4-CI reservoir waters.
into the circulating fluid. When measured at 25 °C, the pH of flashed acid-sulphate water collected at the wellhead may be as low as 2. However, the pH of the water is near neutral at the high temperature in the aquifer. Production of acidity upon cooling is related to the increased acid strength of HSO4− with decreasing temperature (Villa, et al., 2000). The sulphuric acid type fluid is generally formed by any of three kinds of mechanisms, (1) oxidation of H2S, (2) hydrolysis of SO2, or (3) hydrolysis of S (native sulphur) (Matsuda, et al., 2005). The most important difference between the Na-Cl and acid-sulphate waters is that the main pH-buffer of the former is CO2/HCO3−, but HSO4−/SO4−2 in the latter. Compared to Na-Cl waters, acid SO4-Cl waters contain higher concentrations of SO4= and some minor elements, such as Fe and Mg, which are contained in minerals with pH-dependent solubility (Arnorsson and Stefansson, 2007).
One of the earliest published reports on a geothermal well with acid Cl-SO4 waters with excess chloride discharge was in April 1960, in Senkyoro well No. 6 in Japan. Noguchi, et al., (1970) postulated that a small amount of ground water has mixed with a large amount of volcanic gases rising from deep. This reaction produced a liquid with high concentration of corrosive hydrochloric acid underground. Sulfuric acid in the discharge is believed to form from sulfur compound in volcanic gases through oxidation with air. All metals such as iron, aluminum, calcium, magnesium, sodium, manganese, uranium and thorium in the geothermal water are postulated to be derived from the surrounding rocks by the corrosive action of sulfuric acid. Other fields that have reported this type of acidity include Tatun geothermal field (Chen, 1970; White and Truesdell, 1969; Zarrouk, 2004), and in Krafla (Einarrson et al, 2010).
2. DAMAGE TO CARBON STEEL BY ACID GEOTHERMAL FLUID Carbon steel components exposed to geothermal wells discharging acid SO4-Cl type fluids were surface analyzed. A Scanning Electron Microscope (SEM) coupled to an Energy Dispersive X-Ray Spectroscopy (EDS) was used for analyses. Table 1 details the physical and chemical parameters of the fluid environment were the samples were subjected to. The pH of the fluid from the two wells is less than 3. Higher concentration of SO4 is observed in well X6D compared to well X2D but fluid from well X2D is more saline. Both wells are high-temperature with downhole fluid temperature approaching 300°C. The samples were exposed to the fluid at varying duration, the longest being the stinger pipe which was exposed for 43 days. No temperature measurements were made at exposure conditions. However, this can be referenced as saturated steam temperature at the indicated line pressure. The temperature condition in samples exposed to well X2D fluid is about 173°C, while for X6D samples it is estimated at 185°C except for the low pressure samples which is estimated at 140°C. Note that the fluid pH was measured at laboratory condition at around 25°C.
Hydrochloric (HCl) acid in super-heated steam has been reported in the Geyser’s (Truesdell, et al., 1989), Los Humeros (Tello, et al, 2000) and Larderello (D'Amore and Bolognesi, 1994a). Hydrochloric acid (HCl) is transported only in superheated steam because below 300°C it is very soluble in neutral liquid and would have been removed from the vapor if condensate were present. The origin of the HCI in steam dominated fields is related to high-temperature reactions, which mainly forms from boiling of near-neutral NaCl brines or from reaction of halite with silicates (D'Amore and Bolognesi, 1994b). At The Geysers and Larderello, HCl in steam is closely related to the existence of high-temperature zones below the main vapor-dominated reservoir (Walters et al., 1988; D'Amore et al., 1990). In most areas of The Geysers where HCl occurs, steam is sufficiently superheated that the first condensation occurs at the wellhead and corrosion can be prevented by injection of alkaline solutions (e.g. caustic soda). Casings and surface piping of wells with high-CI steam have shown accelerated corrosion and in some cases experiencing failure (Hirtz, et al., 1991)
The cylindrical probes A and C (Figures 1 and 3) were the sensing element of an electrical resistance (ER) monitoring system used to measure the on-line corrosion rates of carbon steel pipeline from the two acidic wells. Sample B in Figure 2 is a section of a 1” by-pass line attached to the branch line of well X2D were acidic fluid was flowing continuously for about 1 month. Metal sample D in Figure 4 is a small piece of a 4”-diameter stinger installed inside well X6D to stabilize a capillary tubing inserted inside the well. Damage to the stinger was extensive with more than 50% material loss due to uniform corrosion and formation of deep cavities and pits throughout the metal surface. A small wedge shaped piece (about 1 cm) of the sample was cut near the pipe thread to examine the corrosion products inside the pipe (Figure 7).
Acid SO4-Cl reservoir fluids are common in arc-type geothermal systems along the circum-Pacific rim (Pang, 2005). Low pH well discharges in the geothermal systems of Japan and the Philippines are more enriched in oxygen-18 or deuterium or both as compared to neutral pH waters found in the same systems such as in Sumikawa, Japan (Ueda, et al., 1991) and Mahanagdong, Philippines (Salonga, et al., 2000) Isotope enrichment of both oxygen-18 and deuterium means higher percentage of magmatic vapor contribution to the well discharges due to the fact that un-neutralized magmatic water carries much abundant hydrogen ions to keep the water at a lower pH (Pang, 2006). Deep acid-sulfate fluids have been encountered in many volcanic geothermal systems, particularly in association with andesitic volcanoes ((A. H. Truesdell et al., 1989); (Arnórsson, Stefánsson, & Bjarnason, 2007; Salonga, Bayon, & Martinez, 2000) and (Matsuda, Shimada, & Kiyota, 2005) . The sulfur species in acid sulfate-chloride thermal water are shown to be volcanic exhalations (Kiyosu and Kurahashi, 1983). The acidity is caused by HCl or HSO4− or both, and evidence indicates that it mostly forms by transfer of HCl and SO2 from the magmatic heat source
Sample E in Figure 5 shows a commercial A106 grade B carbon steel metal coupon that was used to measure corrosion rate in well X6D through mass-loss method. The coupon was inserted into the two-phase line of well X6D using a retractable coupon holder. Note that large pits developed around the surface of the coupon and the edge of the coupon opposite the fluid flow direction was visibly eroded. Crevice corrosion is also evident along a circular outline where the retractable coupon holder washer made of polyethylene was previously attached. 3 New Zealand Geothermal Workshop 2014 Proceedings 24 – 26 November 2014 Auckland, New Zealand
Table 1 Physical and chemical data of the fluid environment where the samples were exposed. Although no definite exposure period is indicated for ER probe in well X2D, the estimate is less than 1 week.
LABEL
pH
Pressure Mpa (g)
Chloride mg/kg
SO4 mg/kg
Exposure Days
Well X2D Electrical Resistance cylindrical probe carbon steel pipe material
A B
2.56 2.56
0.76 0.76
10086 10086
957 957
nd 31
Well X6D Electrical Resistance cylindrical probe carbon steel pipe material Flat strip coupon
C D E
2.47 2.47 2.47
0.28 1.02 1.02
3947 3947 3947
3187 3187 3187
10 43 14
The ER probes (Figures 1 and 3) were cut across the diameter and set in resin and polished. From the cut section (Figure 6), the inner materials such as the wiring and the ceramic filling are clearly visible. The un-corroded metal can be observed in contrast to corrosion products deposited around the outside surface of the probe body. This transition area is examined at low and high magnification in the SEM (Figures 8 and 11). The wedge-shaped section that was cut from the 1”-pipe has a layer of (about 1 mm thick) corrosion products uniformly lining the inner wall. The scanning electron micrographs of the scales deposited are shown in Figure 9. Micrographs of chipped corrosion product from sample D is shown in Figure 10. Surface micrographs of the A106-B grade carbon steel coupon are shown in Figure 12.
found in significant quantity. This may lead to the conclusion that iron-carbonates (as FeCO3) may be present in the sample along with the iron oxides. However, FeCO3 is known to develop typically at higher pH (pH>6) and can redissolve when the solution becomes more acidic (Fajardo, et al., 2013). The composition of the scales in the A106 B corrosion coupon (Table 2) is more distinct than the rest of the samples analysed. The composition of the scale is more varied and includes large amount of zinc and arsenic. These elements are not significant components of A106B type carbon steel and may come from re-crystallized minerals deposited along with the iron oxide. Note that the abundance of silica in the scale may act as a deposition (nucleation) surface for other minerals from the fluid itself. This may explain why other elements are observed in this sample compared to the rest of the samples. As to why there is abundant silica in the scale matrix is not clear at this time.
Analyses of the composition in weight percent of the materials on the surface of the samples are shown in Table 2. Elements that can be identified using EDS are only those that have atomic weights heavier than beryllium. As such, hydrogen, helium and lithium are not included in the quantification (Shindo and Oikawa, 2002). Corrosion product in the samples exposed to X2D fluid that consists (in terms of weight percent) mainly of iron and oxygen with traces of sulphur, silica, and chloride. The absence of significant amounts of silica in the scales indicate the scale is mostly a product of the transformation of the carbon steel material to iron oxides or oxy-hydroxides instead of being re-crystallized from the geothermal fluid. This is in contrast to the observations of Braithwaite and Lichti (1980) and Borshevska et al (1980) of the formation of magnetite layer overlain by an iron-sulfide (pyrite – pyrhottite) matrix observed in Broadlands-Ohaaki fluid. Goldberg and Owen, (1979) noted in their study of corrosion products in geothermal fluid that Fe and Si composition in the scales formed varied from 21% – 57% and 0.6% to 6.7% respectively. This fluctuation in Fe and Si composition was interpreted to be the replacement of Fe in the original corrosion product by Si from the scale. This is thought to be a transition zone between scale and corrosion product.
Figure 1 Carbon steel cylindrical electrical resistance (ER) probe exposed to well X2D fluid. This probe was cut along the diameter and mounted on resin (Figure 6) to view the corrosion behavior on the metal surface.
Note that in the tests conducted by Shoesmith, et al. (1980) on aqueous H2S fluids. The dominant process observed at pH 2.7 is the direct dissolution of iron and only small traces of cubic ferrous sulphide and small traces of mackinawite developed on the metal surface. The appearance of the X2D samples in this case may be similar.
Figure 2 Carbon steel pipe (1”) used to transmit X2D fluid for 1 month at low flow condition. A section of pipe was cut near the thread (Figure 7). The inner surface of the pipe is coated with corrosion product and scale. Deep pits are observed to form around the outer surface of the ER probe (Figures 8 and 11) as well as a general transformation of the metal into corrosion products. The corrosion product appears to be tightly adhering to the surface of the metal due to the absence of space in between
Iron and oxygen also dominate the composition of the scale in the samples exposed to well X6D. No traces of sulfur, or silica were detected from the samples. However, carbon is 4
New Zealand Geothermal Workshop 2014 Proceedings 24 – 26 November 2014 Auckland, New Zealand
them, but the corrosion product is fragile and shows cracks at certain points from where acid fluid can penetrate and react with the metal thereby resulting in continuous material loss. The corrosion rate of carbon steel in well X2D
easily abraded. There is general corrosion observed throughout the surface of the metal as well as deep pits observed in the ER probes (Figure 11). At high magnification, several crystal types are observed in the scale possibly indicating different minerals including iron carbonates in addition to iron oxides. The crystals in sample E (Figure 12) have been damaged due to handling but appear to be similar in sample D (Figure 11). The corrosion rate measured from this coupon is 0.68 mm/y.
Figure 3 The sample above is the remaining piece of a cylindrical ER probe used to monitor well X6D corrosion rate. After 10 days of exposure half of the probe was cut due to corrosion.
Figure 4 A piece of 4” diameter carbon steel pipe used as a capillary tubing stinger inserted into well X6D. More than half of the stinger has been lost due to corrosion before it was retrieved. Figure 6 Cylindrical ER probe cut across the diameter and mounted on resin and polished. Note the wire and ceramic inside the probe. Scale and corrosion products line the outer wall of the probe. Further work is needed to identify the corrosion products beyond chemical composition provided by EDS. Borshevska, et al. (1982) used an x-ray diffractometer to Figure 5 Carbon steel (A106-B Grade) mass loss coupon (flat strip) exposed to well X6D fluid for 14 days. The coupon was inserted inside the two phase pipeline of the well using a retractable coupon holder. Note the large corrosion pits on the surface of the coupon. The coupon edge facing the downstream flow of the fluid is eroded. measured from this probe is 9.3 mm/y. This is higher as expected compared to the corrosion rate of 3.3 mm/y measured by Miranda-Herrera, et al. (2010) for ASTM A53 Grade B carbon steel exposed to geothermal fluids in Cerro Prieto with pH about 4.5. This is because the fluid in well X2D is more acidic and exposure is at a higher temperature. Shannon, et al (1977) observed than corrosion rates increase 3 – 4 times with 1 unit decrease in pH. When viewed at 20,000X magnification, the corrosion products appear as fibre-shaped crystals (Figure 11) possibly iron oxy-hydroxides (FeOOH). Well-formed hexagonal discshaped crystals are found on the inner surface of the 1” pipe. The scale is untouched prior to the analyses being inside the pipe walls which prevented damage to the crystal from rubbing and handling. Further work using x-ray diffractometry (XRD) is needed to determine the exact identity of individual crystals in the scale deposits.
Figure 7 A small section (about 1 cm long) was cut from the pipe in Figure 2 and attached to a stub mount. The inside surface of the pipe is covered with about 1 mm thick corrosion product.
Even at low magnifications, samples C, D and E show extensive damage (Figures 10 -12). The metal has been completely transformed to brittle corrosion products that are 5 New Zealand Geothermal Workshop 2014 Proceedings 24 – 26 November 2014 Auckland, New Zealand
study the crystal patterns of corrosion products and identify individual minerals in the corrosion products. Another method would be to use a Transmission Electron microscope (TEM) in in conjunction with XRD similar to the approach of (Harrar et al., 1979).
number of weeks and are vulnerable to this problem. In subsequent work, surface analyses should be done as soon as the samples are retrieved from the test facilities. 3. CORROSION TESTS USINGS CRA’S IN ACIDIC GEOTHERMAL WELLS A database of various metals and corrosion resistant alloys that have been tested in acidic geothermal fluids or laboratory fluid analogues have been created as part of this study. A total of 87 distinct metals and alloys have been tested in geothermal fluid based on published information.
Handling of the samples from field exposure to the laboratory presents a constraint in preserving the crystals developed during fluid exposure of the samples. Fluid trapped in the mineral scale tends to react with the scale over time at much lower temperature than in the original test
Figure 8 Scanning electron micrograph of the cylindrical ER probe cross-section. Low magnification image (left) show the metal and corrosion product (CP) interface. General corrosion is evident around the metal surface with occasional deep pits. High magnifications image of the pit area (right) shows the texture of the iron-oxygen corrosion product. Cracks developed within the CP matrix allowing the fluid to penetrate deep into the metal surface and continue the corrosion process.
Figure 9 Corrosion product deposited inside the 1” carbon steel pipe used in well X2D. Right shows high magnification image of the iron-oxygen corrosion product with hexagonal plate crystal structure. environment. These samples have been in storage for a
Data were collected from 30 research reports and papers dealing with corrosion tests of metals and alloys in geothermal fluids. The database contains the following 6 New Zealand Geothermal Workshop 2014 Proceedings 24 – 26 November 2014 Auckland, New Zealand
information pH, temperature, chloride concentration and fluid velocity as these were almost always indicated for each test. Although, hydrogen sulfide, sulfate and oxygen concentrations are important from a corrosion standpoint, most of the tests did not specify these parameters.
Although attempts have been made to identify each metal or alloy using the Unified Numbering System (UNS), most authors did not provide this information. Usually the researchers indicate the common or commercial names for the metal or alloy tested with a few exceptions. The UNS number is not a specification, because it does not provide the
Figure 10 Chipped corrosion products from the carbon steel pipe exposed to well X6D fluid. The metal has been completely transformed to brittle corrosion product without any clear crystal pattern (amorphous?) shown at high magnification (right).
Figure 11 Surface micrograph of cylindrical ER probe exposed to X6D fluid. Note the micro-cracks (left) on the surface of the corrosion product. Filament-like crystals are growing from the iron oxide mass seen at higher magnification (right). method of manufacturing from which the material is produced (e.g., pipe, bar, forging, casting, plate) rather it provides the chemical composition of the alloy (Baboian, 2005). From the database, it is noted that the most tested metals are the carbon steels conforming to American Petroleum Institute (API) Specification 5CT or International Standards Organization (ISO) 11960:2004. This standard specifies the steel pipes for use as well casing or tubular for oil and gas 7 New Zealand Geothermal Workshop 2014 Proceedings 24 – 26 November 2014 Auckland, New Zealand
and similarly adopted for geothermal applications. Understandably K55, J55, N80 and L80 carbon steels are well represented in majority of the tests. Also included with this group are the American Iron and Steel Institute (AISI) plain carbon steels 1010, 1018 and 1020 which is compliant with ASTM A519 standard for carbon and alloy steel mechanical tubing. Weathering steel ASTM grades A 588 (COR-TEN® B), grey cast iron, P110 and P235GH (European Standard equivalent to ASTM A106 Grade A carbon steel) has been likewise tested in acid geothermal fluid condition.
corrosion cracking. Also within the austenitic group tested in acid geothermal fluid in this database are AL6XN, 904L, 1925hMo, 310, Sanicro 28 (high-alloy austenitic) and 317LM. Ferritic stainless steels are resistant to chloride stress corrosion cracking, and have high strength and are highly resistant to vibration. Those tested in acidic geothermal fluid in this summary are 405, 430, 444 and E-Brite. This type of steel is known to be subject to 885ºF (475ºC) embrittlement at temperatures as low as 600ºF (315ºC) and susceptible to hydrogen embrittlement. Martensitic stainless steels tested include 410 and 422. Martensitic stainless steels are not highly corrosion resistant and only as good as Type 304.
Kurata, et al. (1995) use Japanese made chromiummolybdenum (Cr-Mo) steels STBA23, STBA24, STBA25
Figure 12 Minute craters (red circles) form on the surface of the corrosion coupon (left). These eventually grow in size and become large crevices seen in Figure 5. Iron oxide crystal formed at the surface of the coupon is seen at higher magnification. and STBA 26 that conforms to Japan standard (JIS G3462) for most of their tests. The equivalent ASTM standard for these alloys is A213 or 10216-2 for the European Standard (EN).
However, these stainless steels have excellent hardness from the carbon added to the alloy. Their ability to maintain a keen edge comes from their high hardness and corrosion resistance. Although SUS630 (equivalent to 17-4PH), is a martensitic stainless steel, it is classified in the precipitation hardening stainless steels category. SUS630 is an ideal metal to use when high strength and corrosion resistance is required.
Also thoroughly tested in acidic geothermal fluid are the austenitic stainless steel Types 304, 316 and 316L. These are the most popular of the stainless steels because of their ductility, ease of working and good corrosion resistance. These alloys are also not subject to 885ºF (475ºC)
Alloys 22015, 2507 and SUS329J2L belong to the duplex, types of stainless steel. Duplex stainless steels are characterized by having both austenite and ferrite in their microstructure. Duplex stainless steels are susceptible to 885ºF (475ºC) embrittlement at temperatures as low as 600ºF (315ºC) and to hydrogen embrittlement. The latter is resistant to chloride stress corrosion cracking.
Table 2 Results of analyses of the metal samples surface composition using energy dispersive x-ray analysis (EDS). A
B C D Composition (in % wt)
30.41 0.52 1.55 0.19 67.33
26.57 0.1 0.22 1.75 71.36
100
100
Element C O Si S Cl Fe Al Zn As Total
E
6.27 46.43
4.34 16.94
0.76 46.53
1 77.71
100
100
6.54 18.67 19.26 8.51
Nickel-based alloys were also substantially tested in acid geothermal fluid. The following are included in the summary: Monel 400, C-276, 625, 825, Alloy 59, Inconel 600, Alloy 20 (Carpenter 20Cb3), Alloy 31, G3. Nickel based alloys are very resistant to corrosion in alkaline environments, neutral chemicals and many natural environments. In addition, many nickel based alloys show excellent resistance to pitting, crevice corrosion and stress corrosion cracking in chloride environments. These alloys however are more expensive than stainless steels (Sorell, 1998) due to high nickel content.
19.68 0.52 12.83 13.99 100
embrittlement and to hydrogen embrittlement. However, these are known to be susceptible to chloride stress 8
New Zealand Geothermal Workshop 2014 Proceedings 24 – 26 November 2014 Auckland, New Zealand
Another group that is thoroughly tested in acidic geothermal brine is the titanium and titanium alloys (Thomas, 2003). In the database, the following alloys have been tested in addition to pure titanium: Ti-6Al-4V, Titanium Grade 1 and Titanium Grade 2. Other pure metals tested are niobium, nickel, molybdenum and some specialty aluminum (Aluminum 5083) and cobalt (S 816, Haynes 25) alloys.
The actual chemical composition in weight percent of each alloy as provided by the manufacturers is also shown in the table. Stainless steels 2205 and 2507 are duplex and superduplex respectively. Classification of standard, super or hyper duplex stainless steels depends on the Pitting Resistance Equivalent Number (PREN). The PREN can be calculated using the equation:
Corrosion rates were measured using mass-loss coupon exposure method, linear polarization resistance (LPR) technique and potentiodynamic polarization (mainly for localized corrosion such as pitting and crevice corrosion). By far, the most prevalent and the method used to calibrate other measurements are through mass-loss coupon. The authors used different units to measure corrosion rate and no attempt was made in the database to reckon the values to one standard unit. Conversions of the units are straightforward except when using mdd units where density data need to be available to do conversion to mils/y or µm/y. Acceptable corrosion rates for metals in geothermal well environment is 100 m/s at pH ≤ 3.2 (Sanada, et al., 1998). Turbulence of the fluid flow also increases the corrosion rates of the test alloys at lower pH and higher temperature. Metals that corroded or have marginally acceptable corrosion rates at static condition in a given geothermal fluid temperature and pH level are expected to have higher corrosion rates at higher fluid velocities.
PREN = %Cr + 3.3(%Mo) + 16(%N)
(5)
Standard duplex stainless steels have PREN from 30-40, super-duplex > 40, and hyper-duplex >47. PREN values of 2205 and 2507 are 36 and 42 respectively. A PREN of 32 is considered the minimum for the prevention of seawater pitting resistance. Both alloys have been tested in acidic geothermal environment (Kurata, et al, 1995, Lichti et al, 2010, Sanada, et al, 2000, and Kurata, et a., 1992) and gave acceptable (3.0. AL6XN alloy is a super-austenitic stainless steel developed by Allegheny Ludlum Corporation (Malik, et al., 2000). This metal has been tested in RRGE-1 well in Raft River by Miller (1977) and in Magmamax-1 geothermal tests facility in Salton Sea (Carter and McCawley, 1982). Although austenitic stainless steels do not perform well in chloride solutions due to pitting corrosion, the measured general corrosion rate of 6X is < 5 µm/y (Carter and MacCawley, 1982). These tests were carried out at high temperature and in hyper-saline condition. However, this alloy has not been tested yet in pH < 5.2 and maybe worthwhile to test in the low pH range, especially the improved AL6XN type. Alloys I625, I825, Alloy 59 and Alloy 31 are nickel-based alloys with varying nickel and molybdenum content (Sorell, 1997). These alloys have been substantially tested in high temperature, high salinity and low-pH conditions. Normally, higher nickel and molybdenum content provide better corrosion resistance. Kurata, et al., (1995) concluded that for geothermal fluid at temperature of 300°C and pH 3.0, the suitable material for well construction should have more than 30% (weight) of Cr equivalent and corresponds to duplex stainless steel or an equivalent high Ni - based alloy. Alloy 59 is suitable and represents a safe option to be used in geothermal facilities working with highly saline fluids up to 150 °C (Mundhenk et al., 2013).
The mild carbon steels (including API Spec 5CT materials) are not suitable for acidic geothermal fluids with corrosion rates exceeding the threshold value at pH 3 consistent with the data from Kurata, et al, (1992) and Shannon (1977). Schutz and Watkins, (1998) concludes that ruthenium-enhanced titanium alloys “may be the only tubular materials fully resistant enough for development of high-temperature (
